KR101685419B1 - Apparatus and method for processing digital image - Google Patents

Abstract

The present invention relates to a digital image processing apparatus and method for performing image processing using an image sensor. According to an embodiment of the present invention, there is provided a digital image processing method comprising the steps of: (a) detecting motion of an input image; (b) normalizing a color signal of the detected motion image using the luminance signal of the detected motion image; (c) detecting skin from the normalized signal using a back propagation algorithm; And (d) zooming around the region detected as the skin. According to the present invention, the recognition rate of the neural network skin color is increased by using the normalized skin color data, thereby increasing the recognition rate according to the illuminance and reducing the amount of calculation and learning time, thereby enabling fast zoom processing.

Description

[0001] Apparatus and method for processing digital image [

The present invention relates to a digital image processing apparatus and method for performing image processing using an image sensor.

In the case of motion in a general surveillance system, the zoom was activated in the camera according to the movement. Especially, in case of motion, zooming of the camera by zooming the moving object is often used to monitor the object in more detail.

Conventional surveillance systems often use smart zoom according to motion only in consideration of motion, but there is also a problem in that the movement of objects due to other influences such as climatic influences (wind, rain) also operates. Particularly, in general, it is often necessary to monitor a person, especially a face, of a moving object. In this case, the smart zoom is easily operated even for movement of an object other than a person. In addition, existing face recognition technology has a problem that it is not easy to embed an algorithm in a surveillance camera system having a small size because of a long time and a complicated size of execution time. Therefore, a simple technique for monitoring only a human face region in real time when a motion occurs is required.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a digital image processing apparatus and method for performing zooming using normalized skin color data in order to increase the recognition rate according to illumination.

According to an embodiment of the present invention, there is provided a digital image processing method comprising the steps of: (a) detecting motion of an input image; (b) normalizing a color signal of the detected motion image using the luminance signal of the detected motion image; (c) detecting skin from the normalized signal using a back propagation algorithm; And (d) zooming around the region detected as the skin.

In the present invention, a normal signal may be output by dividing a value obtained by adding the first normalization factor to the color signal at the time of normalization in the step (b) by the luminance signal, and then multiplying the luminance signal by a second normalization factor.

In the present invention, the first normalization factor is a bit shift factor that makes the color signal positive, and the second normalization factor may be a value between 0 and 1 to generate the backpropagation algorithm input value .

According to another aspect of the present invention, there is provided a digital image processing apparatus including: a motion detector for detecting motion using an input video signal and a previous video signal; A normalization unit for normalizing a color signal of the detected motion image using the luminance signal of the detected motion image; A skin detector for detecting skin from the normalized signal using a back propagation algorithm; And a zoom controller for outputting a zooming signal around the area detected as the skin.

In the present invention, the normalizing unit may divide a value obtained by adding the first normalization factor to the color signal into the luminance signal, and then output the normal signal by multiplying the luminance signal by a second normalization factor.

In the present invention, the first normalization factor is a bit shift factor that makes the color signal positive, and the second normalization factor may be a value between 0 and 1 to generate the backpropagation algorithm input value .

As described above, according to the present invention, the recognition rate of the neural network skin color is increased by using the normalized skin color data, thereby increasing the recognition rate according to the illuminance and reducing the amount of calculation and learning time to enable fast zoom processing .

1 is a diagram illustrating a monitoring system to which a digital image processing apparatus according to an embodiment of the present invention is applied.
2 is a block diagram showing a configuration of a digital image processing apparatus according to an embodiment of the present invention.
3 is a detailed block diagram of the digital signal processing unit in FIG.
FIG. 4 is a view for explaining weight values generated by prior learning using a back propagation algorithm in FIG.
5 is a flowchart illustrating an operation of a digital image processing method according to another embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention is capable of various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The terms first, second, etc. may be used to describe various elements, but the elements should not be limited by terms. Terms are used only for the purpose of distinguishing one component from another.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. Referring to the accompanying drawings, the same or corresponding components are denoted by the same reference numerals, do.

1 is a diagram illustrating a monitoring system to which a digital image processing apparatus according to an embodiment of the present invention is applied.

1, the surveillance cameras 101, 111, and 121 communicate with the host apparatus 2 as a host apparatus by communication signals Dcom and generate a live view video signal Svid, To the host device (2). The video signal Svid received from the host device 2 is displayed through a display device while being stored in a recording device, for example, a hard disk drive.

Each of the surveillance cameras 101, 111 and 121 transmits the video signal Svid to the host apparatus 2 while communicating with the host apparatus 2 via a coaxial cable. Accordingly, the communication signals Dcom are transmitted and received in the vertical blank interval of the video signal Svid transmitted to the host apparatus 2. [

Each of the surveillance cameras 101, 111, and 121 performs panning and tilting of the left and right rotations according to the control signals from the host apparatus 2. [

Here, the host apparatus 2 controls the operation of each of the surveillance cameras 101, 111, and 121 according to the surveillance camera control method according to an embodiment of the present invention. The related contents will be described in detail with reference to Figs. 2 to 4. Fig.

2 is a block diagram showing a configuration of a digital image processing apparatus according to an embodiment of the present invention.

Referring to FIG. 2, the digital image processing apparatus includes an optical system (OPS), a photoelectric conversion unit (OEC), a correlation double sampler and an analog-to-digital converter (CDS-ADC) 210, a timing circuit 220, A digital signal processor 230, a video-signal generator 240, a diaphragm motor M _A , a zoom motor M _Z , a focus motor M _F , a filter motor M _D , A motor M _p , a tilting motor M _T , a driver 250, a communication interface 260, a micro-computer 270 and an illumination unit 280.

An optical system (OPS) including a lens unit and a filter unit optically processes light from a subject.

The lens portion of the optical system OPS includes a zoom lens and a focus lens. In the filter section of the optical system (OPS), an optical low pass filter (OLPF) used in the night operation mode removes high frequency content of optical noise. Infrared cut filter (IRF) used in the daytime operation mode blocks the infrared component of incident light.

A photoelectric conversion unit (OEC) of a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) converts light from an optical system (OPS) into an electrical analog signal. The digital signal processor 230 controls the timing circuit 220 to control the operation of the photoelectric conversion unit OEC and the correlation double sampler and analog-to-digital converter 210.

The CDS-ADC 210 processes the analog video signal from the photoelectric conversion unit OEC, removes the high-frequency noise, adjusts the amplitude, and converts the digital video data. The digital image data is input to the digital signal processor 230.

A digital signal processor 230 as a main control unit processes digital signals from the CDS-ADC device 210 to generate digital image data classified into luminance and chrominance signals.

The video signal generator 240 converts the digital image data from the digital signal processor 230 into a video signal S _VID which is an analog video signal.

A digital signal processor 230 as a main control unit includes, for the host device, for example, via the communication interface 260 and the communication channel (D _COM), the video and communicates with the computer through the video signal channel (S _VID) - signal And transmits the video signal from the generating unit 240 to the host devices.

The microcomputer 270 controls the driving unit 250 to drive the diaphragm motor M _A , the zoom motor M _Z , the focus motor M _F , the filter motor M _D , the panning motor M _p , And the tilting motor M _T. In addition, the micro-computer 270 controls the illumination unit 280 to illuminate the illumination light.

The diaphragm motor M _A drives a diaphragm, the zoom motor M _Z drives a zoom lens, and the focus motor M _F drives a focus lens. The filter motor M _D drives the optical low-pass filter OLPF and the infrared cut filter IRF in the filter unit.

The panning motor M _P rotates the optical system OPS to the left and right. The tilting motor M _T rotates the optical system OPS up and down.

Hereinafter, the recognition of the neural network skin color using the normalized skin color data performed by the digital signal processor 230 as the main control unit will be described with reference to FIGS. 3 and 4. FIG.

3 is a detailed block diagram of the digital signal processor 230 in FIG.

3, the digital signal processing unit 230 includes a buffer 231, a motion detection boom 232, a conversion unit 233, a normalization unit 234, a skin detection unit 235, a position calculation unit 236, And a zoom control unit 237.

The motion detector 232 compares the input current (t) image and the previous (t-1) image stored in the buffer 231 to detect motion.

The conversion unit 233 converts the RGB signal of the image in which the motion is detected into a luminance signal and a YCbCr signal which is a color difference signal.

The normalization unit 234 normalizes the color signal CbCr of the detected motion image using the luminance signal Y of the detected motion image, and the normalized output signal DbDr is expressed by Equation 1 below.

In Equation (1), n is a first normalization factor, and is a value for shifting a bit to represent a positive value when a Cr value is represented by a negative number. And m is a coefficient between 0 and 1 to generate an input value when using a back propagation algorithm for skin detection described later with a second normalization factor. Since the color signal CbCr is divided into the luminance signal Y at the time of normalization in this way, the influence of the illuminance is removed at the time of skin detection using the back propagation algorithm, the judgment boundary region is small and the skin recognition rate can be increased, The amount of calculation and the learning time are reduced, and the processing speed is improved.

The skin detecting unit 225 detects the skin by performing a back propagation algorithm with a weight obtained by inputting the normalization signal DrDb and pre-learning.

Fig. 4 is a diagram for explaining weight values generated by prior learning using a back propagation algorithm. Referring to FIG. 4, the backpropagation algorithm performing unit 235-1 is composed of a multilayer of input layer / hidden layer / output layer, which is useful in setting a nonlinear boundary region. Generally, input and output data are required for learning. The input data is a normalized value (DbDr, see FIG. 3) obtained by converting the RBG signal of the input image into the YCbCr signal in advance. When the input data DrDb is multiplied by the weight (W _ji , W _kj ) of the neural network and the result is output, the error is calculated by comparing with the desired output value and the weight of the output layer and the hidden layer through the process of updating the _ji, W _kj) the station to update the weight (W _ji, W _kj). The output data proceeds in the direction of the input layer, the hidden layer, and the output layer, and the updating of the weights w _ji and w _kj proceeds in the direction opposite to the output layer and the hidden layer. Since such a back propagation algorithm is a well-known technique, a detailed description thereof will be omitted.

The skin detecting unit 235 performs a back propagation algorithm and outputs 1 if the skin color is detected in the corresponding image signal, and 0 otherwise.

The position calculation unit 236 counts the number of pixels detected by the skin, and calculates its position.

If the number of pixels counted by the position calculation unit 236 is equal to or greater than a predetermined number, the zoom control unit 237 outputs a zoom control signal for zooming the position to the driving unit 250. The driving unit 250 receives the zoom control signal and operates the zoom motor M _Z.

5 is a flowchart illustrating an operation of a digital image processing method according to another embodiment of the present invention.

The digital signal processing unit 230 detects motion by comparing the current (t) image inputted and the previous (t-1) image stored in the buffer 231 (step 501).

Then, the digital signal processor 230 converts the RGB signal of the motion-detected image into a YCbCr signal (step 503).

The digital signal processor 230 normalizes the color signal CbCr using the luminance signal Y, the first normalization factor, and the second normalization factor (step 505). The normalized signal DbDr is used as input data of the back propagation algorithm.

Then, the digital signal processor 23 detects the skin by performing a back propagation algorithm with a weight obtained by inputting the normalization signal DrDb and pre-learning (step 507). When the skin color is detected, the digital signal processing unit 230 outputs 1 if it is detected, and 0 otherwise.

When the skin color is detected, the digital signal processor 230 counts the number of pixels detected by the skin and calculates its position (step 509).

When the position is calculated, the digital signal processor 230 outputs a zoom control signal for zooming around the position to the driving unit 250. The driving unit 250 receives the zoom control signal and outputs the zoom control signal to the zoom motor M _Z , (Step 511).

The specific acts described in the present invention are, by way of example, not intended to limit the scope of the invention in any way. For brevity of description, descriptions of conventional electronic configurations, control systems, software, and other functional aspects of such systems may be omitted. Also, the connections or connecting members of the lines between the components shown in the figures are illustrative of functional connections and / or physical or circuit connections, which may be replaced or additionally provided by a variety of functional connections, physical Connection, or circuit connections. Also, unless stated otherwise such as " essential ", " importantly ", etc., it may not be a necessary component for application of the present invention.

The use of the terms " above " and similar indication words in the specification of the present invention (particularly in the claims) may refer to both singular and plural. In addition, in the present invention, when a range is described, it includes the invention to which the individual values belonging to the above range are applied (unless there is contradiction thereto), and each individual value constituting the above range is described in the detailed description of the invention The same. Finally, the steps may be performed in any suitable order, unless explicitly stated or contrary to the description of the steps constituting the method according to the invention. The present invention is not necessarily limited to the order of description of the above steps. The use of all examples or exemplary language (e.g., etc.) in this invention is for the purpose of describing the present invention only in detail and is not to be limited by the scope of the claims, It is not. It will also be appreciated by those skilled in the art that various modifications, combinations, and alterations may be made depending on design criteria and factors within the scope of the appended claims or equivalents thereof.

101, 111, 121: Surveillance camera 2: Host device
OPS: Optical system OEC: Photoelectric conversion unit
210: CDS-ADC 220: Timing circuit
230: digital signal processor 240: video-signal generator
250: driving unit 260: communication interface
270: micro-computer 280: illuminator
231: buffer 232: motion detector
233: conversion unit 234: normalization unit
235: skin detecting unit 236: position calculating unit
237:

Claims (6)

(a) detecting a motion of an input image;
(b) normalizing a color signal of the detected motion image using the luminance signal of the detected motion image;
(c) detecting skin from the normalized signal using a back propagation algorithm; And
(d) zooming around a region detected as the skin,
At the time of normalization in the step (b)
Dividing a value obtained by adding the first normalization factor to the color signal into the luminance signal, multiplying the luminance signal by a second normalization factor, and outputting the normalized signal,
Wherein the first normalization factor comprises a bit shift factor that causes the color signal to be a positive value and wherein the second normalization factor comprises a digital shift value that includes a value between 0 and 1 to generate an input value of the back propagation algorithm, Image processing method. delete delete A motion detector for detecting motion using an input video signal and a previous video signal;
A normalization unit for normalizing a color signal of the detected motion image using the luminance signal of the detected motion image;
A skin detector for detecting skin from the normalized signal using a back propagation algorithm; And
And a zoom controller for outputting a zooming signal around an area detected by the skin,
The normalization unit may include:
Dividing a value obtained by adding the first normalization factor to the color signal into the luminance signal, multiplying the luminance signal by a second normalization factor, and outputting the normalized signal,
Wherein the first normalization factor comprises a bit shift factor that causes the color signal to be a positive value and wherein the second normalization factor comprises a digital shift value that includes a value between 0 and 1 to generate an input value of the back propagation algorithm, Image processing apparatus. delete delete

Images